Presented by NinjaOne & Carahsoft
Artificial intelligence is increasing the speed and scale of cyber operations for both defenders and adversaries. Egon Rinderer, Senior Vice President of Federal and Enterprise Growth at NinjaOne, says government should view that competition as an arms race—and recognize that its opponents may operate without the same safeguards or restrictions.
Speaking at the Billington Cybersecurity Summit 2026, Rinderer explains that the pace of AI development makes the current competition different from previous technology cycles. Adversaries can use the technology to find vulnerabilities, generate convincing attacks and operate across large numbers of targets more quickly than human operators working alone.
The government and private sector, meanwhile, must place appropriate controls around powerful capabilities. Those safeguards are important, but they create an inherent imbalance when nation-state adversaries impose few or no equivalent restrictions on their operators.
“How does that arms race work when we are willing to put those safeguards in place and hamstring ourselves to some degree, and our adversary is not willing to do that?” Rinderer asks.
He does not claim there is a simple answer. The question instead highlights the need to treat AI-enabled cyber activity as a strategic risk rather than another incremental improvement to existing technology.
When a vendor releases a patch for a critical vulnerability, agencies need to test and deploy it across every affected device. That process often includes several manual steps and handoffs. Each one consumes time while the vulnerability remains available to attackers.
Requirements to remediate known exploited vulnerabilities can give organizations only a short period to respond. Rinderer says many enterprises would struggle to deploy a critical patch throughout their entire environment within 72 hours.
AI can help compress the timeline between the release of a patch and its application across the enterprise. It can assist with identifying affected devices, evaluating potential compatibility issues, scheduling deployments and verifying that remediation was completed.
The objective is not to remove people from the process entirely. It is to reduce the amount of routine work that depends on human intervention so specialists can concentrate on exceptions, operational risks and decisions that require judgment.
Rinderer compares the impact of AI to earlier computing advances. A computer was once a person who performed calculations. The machines that adopted the name became powerful because they could complete those calculations at a speed humans could not match.
AI can provide a similar force multiplier. It allows a person or team to examine more information, complete repetitive tasks and respond to events more quickly.
That speed matters because attackers can also automate their work. An adversary may use AI to identify exposed systems, customize phishing messages or modify an attack in response to a defender’s actions.
Government cannot rely on human-scale processes when threats are operating at machine speed. Agencies must use automation to reduce the time required to detect, analyze and remediate vulnerabilities without sacrificing the controls needed for responsible use.
Speed may not ultimately be the most consequential change. Rinderer believes the greater turning point will come when AI can at least approximately simulate reasoning.
That ability would allow systems to move beyond executing predefined tasks and make more complex decisions based on changing circumstances. For defenders, that could improve investigation, prioritization and response. An AI system might connect seemingly unrelated events, identify a developing attack and recommend action before a human team reaches the same conclusion.
The prospect is also unsettling because adversaries would have access to comparable capabilities. An offensive system capable of reasoning through obstacles could adapt its approach when a defense blocks its initial path.
Rinderer describes that future development as both powerful and frightening. The government therefore needs to advance its defensive uses of AI while continuing to address security, oversight and ethical risks.
The most practical place to begin is with tangible problems. Faster patching, improved asset management and automated routine work may not receive the attention of more futuristic AI capabilities, but they can reduce exposure today.
The cyber arms race will not be won by adopting AI for its own sake. Agencies need to identify where the technology measurably improves their security posture and deploy it in ways that help defenders keep pace with rapidly advancing threats.
Key Takeaways